U.S. patent number 11,176,321 [Application Number 16/401,254] was granted by the patent office on 2021-11-16 for automated feedback in online language exercises.
This patent grant is currently assigned to International Business Machines Corporation. The grantee listed for this patent is International Business Machines Corporation. Invention is credited to Ismail Yunus Akhalwaya, Toby Kurien, Maletsabisa Molapo, Richard Allen Young.
United States Patent |
11,176,321 |
Kurien , et al. |
November 16, 2021 |
Automated feedback in online language exercises
Abstract
Language models may be run with an input set of words in a given
sentence. Each of the language models can predict a set of next
candidate words to follow the input set or words. Based on the sets
of next candidate words predicted by the language models and an
actual next word, language guidance can be provided.
Inventors: |
Kurien; Toby (Midrand,
ZA), Young; Richard Allen (Johannesburg,
ZA), Molapo; Maletsabisa (Pretoria, ZA),
Akhalwaya; Ismail Yunus (Emmarentia, ZA) |
Applicant: |
Name |
City |
State |
Country |
Type |
International Business Machines Corporation |
Armonk |
NY |
US |
|
|
Assignee: |
International Business Machines
Corporation (Armonk, NY)
|
Family
ID: |
1000005934677 |
Appl.
No.: |
16/401,254 |
Filed: |
May 2, 2019 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20200349224 A1 |
Nov 5, 2020 |
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G06F
40/232 (20200101); G06F 40/274 (20200101); G06N
3/0454 (20130101); G09B 19/06 (20130101); G06N
3/08 (20130101) |
Current International
Class: |
G06F
40/232 (20200101); G09B 19/06 (20060101); G06N
3/08 (20060101); G06N 3/04 (20060101); G06F
40/274 (20200101) |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
108519974 |
|
Sep 2018 |
|
CN |
|
108681533 |
|
Oct 2018 |
|
CN |
|
2019/024050 |
|
Feb 2019 |
|
WO |
|
Other References
NIST, "NIST Cloud Computing Program",
http://csrc.nist.gov/groups/SNS/cloud-computing/index.html, Created
Dec. 1, 2016, Updated Oct. 6, 2017, 9 pages. cited by
applicant.
|
Primary Examiner: Maung; Thomas H
Attorney, Agent or Firm: Scully, Scott, Murphy &
Presser, P.C. Benjamin; Shimon
Claims
What is claimed is:
1. A system comprising: at least one hardware processor; a memory
device coupled to the at least one hardware processor; a first
language model stored on the memory device; and a second language
model stored on the memory device; the at least one hardware
processor operable to at least: run the first language model with
an input set of words in a given sentence, the first language model
outputting a first set of candidate words predicted to follow the
input set of words in the given sentence, the first language model
further outputting a score associated with each of the candidate
words in the first set of candidate words; run the second language
model with the input set of words in the given sentence, the second
language model outputting a second set of candidate words predicted
to follow the input set of words in the given sentence, the second
language model further outputting a score associated with each of
the candidate words in the second set of candidate words; receive
an actual word following the input set of words; responsive to
determining that the actual word matches with a candidate word in
the first set of candidate words, update a first cumulative tally
associated with the first language model with the score associated
with the candidate word in the first set matching the actual word;
responsive to determining that the actual word matches with a
candidate word in the second set of candidate words, update a
second cumulative tally associated with the second language model
with the score associated with the candidate word in the second set
matching the actual word; compare the first cumulative tally with
the second cumulative tally; responsive to determining that the
first cumulative tally and the second cumulative tally deviate from
one another by more than a pre-defined threshold, identify the
actual word in the given sentence for flagging; and keep track of a
position in the given sentence where the first cumulative tally
deviated from the second cumulative tally, wherein the first
language model is trained based on a first training data set
including at least data determined to have proper usage of a
language, and the second language model is trained based on a
second training data set including at least data determined to have
incorrect usage of the language.
2. The system of claim 1, wherein the first language model and the
second language model are artificial neural network models.
3. The system of claim 2, wherein the first language model and the
second language model are recurrent neural network models.
4. The system of claim 1, wherein the first training data set
includes at least data associated with a category of submission,
wherein the first language model is trained to output the first set
of next words likely to be associated with language found in the
category of submission.
5. The system of claim 1, wherein the first training data set
includes at least data associated with a category of submission
grouped by a cohort, wherein the first language model is trained to
output the first set of next words likely to be associated with
language found in the category of submission grouped by the
cohort.
6. The system of claim 1, wherein the given sentence is fed into
the first language model and the second language model one word at
a time.
7. The system of claim 1, wherein the at least one hardware
processor is operable to cause flagging of the actual word in the
given sentence.
8. The system of claim 1, wherein the at least one hardware
processor is operable to cause highlighting of the actual word in
the given sentence.
9. The system of claim 1, wherein the at least one hardware
processor is operable to provide the first set of candidate
words.
10. The system of claim 9, wherein the at least on hardware
processor is operable to cause a presentation of the first set of
candidate words.
11. A method comprising: inputting an input set of words in a given
sentence to a first language model, the first language model
outputting a first set of candidate words predicted to follow the
input set of words in the given sentence, the first language model
further outputting a score associated with each of the candidate
words in the first set of candidate words; inputting the input set
of words in the given sentence to a second language model, the
second language model outputting a second set of candidate words
predicted to follow the input set of words in the given sentence,
the second language model further outputting a score associated
with each of the candidate words in the second set of candidate
words; receiving an actual word following the input set of words;
responsive to determining that the actual word matches with a
candidate word in the first set of candidate words, updating a
first cumulative tally associated with the first language model
with the score associated with the candidate word in the first set
matching the actual word; responsive to determining that the actual
word matches with a candidate word in the second set of candidate
words, updating a second cumulative tally associated with the
second language model with the score associated with the candidate
word in the second set matching the actual word; compare the first
cumulative tally with the second cumulative tally; responsive to
determining that the first cumulative tally and the second
cumulative tally deviate from one another by more than a
pre-defined threshold, causing the actual word in the given
sentence to be flagged; and keep track of a position in the given
sentence where the first cumulative tally deviated from the second
cumulative tally, wherein the first language model is trained based
on a first training data set including at least data determined to
have proper usage of a language, and the second language model is
trained based on a second training data set including at least data
determined to have incorrect usage of the language.
12. The method of claim 11, wherein the first language model and
the second language model are artificial neural network models.
13. The method of claim 12, wherein the first language model and
the second language model are recurrent neural network models.
14. The method of claim 11, wherein the first training data set
includes at least data associated with a category of submission,
wherein the first language model is trained to output the first set
of next words likely to be associated with language found in the
category of submission.
15. The method of claim 11, wherein the first training data set
includes at least data associated with a category of submission
grouped by a cohort, wherein the first language model is trained to
output the first set of next words likely to be associated with
language found in the category of submission grouped by the
cohort.
16. The method of claim 11, wherein the given sentence is fed into
the first language model and the second language model one word at
a time.
17. The method of claim 11, wherein the method further comprises
causing a presentation of the first set of candidate words
responsive to the flagged actual word being selected.
18. A computer program product comprising a computer readable
storage medium having program instructions embodied therewith, the
program instructions readable by a device to cause the device to:
run by the device, a first language model with an input set of
words in a given sentence, the first language model outputting a
first set of candidate words predicted to follow the input set of
words in the given sentence; run by the device, a second language
model with the input set of words in the given sentence, the second
language model outputting a second set of candidate words predicted
to follow the input set of words in the given sentence; and based
on the first set of candidate words, the second set of candidate
words, and an actual next word following the input set of words,
provide by the device, guidance for phrasing the given sentence,
keep track of a position in the given sentence where a first
cumulative tally associated the first language model deviate from a
second cumulative tally associated the second language model,
wherein at least one of the first cumulative tally and the second
cumulative tally is updated based on the actual next word, wherein
the first language model is trained based on a first training data
set including at least data determined to have proper usage of a
language, and the second language model is trained based on a
second training data set including at least data determined to have
incorrect usage of the language.
Description
BACKGROUND
The present application relates generally to computers and computer
applications, and more particularly to artificial intelligence,
machine learning and providing automated feedback in language
usage.
There are various tools such as proofing tools and word processing
tools, which may provide or suggest spelling and/or grammatical
error corrections in spoken or written language, such as in English
language. For instance, such tools may be enabled to recognize and
tag words or sentences perceived or determined by the tools to be
incorrectly used.
BRIEF SUMMARY
Systems and methods may be provided, which may provide feedback
automatically in language usage, for example, in online language
exercises. A system, in one aspect, can include at least one
hardware processor. A memory device can be coupled to the at least
one hardware processor. A first language model can be stored on the
memory device and a second language model can be stored on the
memory device. At least one hardware processor may be operable to
run the first language model with an input set of words in a given
sentence. The first language model can output a first set of
candidate words predicted to follow the input set of words in the
given sentence. The first language model can further output a score
associated with each of the candidate words in the first set of
candidate words. At least one hardware processor may be further
operable to run the second language model with the input set of
words in the given sentence. The second language model can output a
second set of candidate words predicted to follow the input set of
words in the given sentence. The second language model can further
output a score associated with each of the candidate words in the
second set of candidate words. At least one hardware processor may
be further operable to receive an actual word following the input
set of words. Responsive to determining that the actual word
matches with a candidate word in the first set of candidate words,
at least one hardware processor may be further operable to update a
first cumulative tally associated with the first language model
with the score associated with the candidate word in the first set
matching the actual word. Responsive to determining that the actual
word matches with a candidate word in the second set of candidate
words, at least one hardware processor may be further operable to
update a second cumulative tally associated with the second
language model with the score associated with the candidate word in
the second set matching the actual word. Responsive to determining
that the first cumulative tally and the second cumulative tally
deviate by more than a pre-defined threshold, at least one hardware
processor may be further operable to identify the actual word in
the given sentence for flagging.
A method, in one aspect, may include running a first language model
with an input set of words in a given sentence, the first language
model outputting a first set of candidate words predicted to follow
the input set of words in the given sentence. The method may also
include running a second language model with the input set of words
in the given sentence, the second language model outputting a
second set of candidate words predicted to follow the input set of
words in the given sentence. The method may further include, based
on the first set of candidate words, the second set of candidate
words, and an actual next word following the input set of words,
providing guidance for phrasing the given sentence.
A method, in another aspect, may include inputting an input set of
words in a given sentence to a first language model. The first
language model can output a first set of candidate words predicted
to follow the input set of words in the given sentence. The first
language model can further output a score associated with each of
the candidate words in the first set of candidate words. The method
may also include inputting the input set of words in the given
sentence to a second language model. The second language model can
output a second set of candidate words predicted to follow the
input set of words in the given sentence. The second language model
can further output a score associated with each of the candidate
words in the second set of candidate words. The method may also
include receiving an actual word following the input set of words.
The method may further include, responsive to determining that the
actual word matches with a candidate word in the first set of
candidate words, updating a first cumulative tally associated with
the first language model with the score associated with the
candidate word in the first set matching the actual word. The
method may also include, responsive to determining that the actual
word matches with a candidate word in the second set of candidate
words, updating a second cumulative tally associated with the
second language model with the score associated with the candidate
word in the second set matching the actual word. The method may
further include, responsive to determining that the first
cumulative tally and the second cumulative tally deviate by more
than a pre-defined threshold, causing the actual word in the given
sentence to be flagged.
A computer readable storage medium storing a program of
instructions executable by a machine to perform one or more methods
described herein also may be provided.
Further features as well as the structure and operation of various
embodiments are described in detail below with reference to the
accompanying drawings. In the drawings, like reference numbers
indicate identical or functionally similar elements.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a diagram illustrating a sentence fed into two models and
processing of outputs of the models in one embodiment.
FIG. 2 shows an example of a user interface display showing a
sentence in one embodiment.
FIG. 3 is an example neural network model in one embodiment, which
can be trained based on a training data set.
FIG. 4 is a flow diagram illustrating a method in one embodiment,
which can provide automated corrective feedback.
FIG. 5 is a flow diagram illustrating a method in another
embodiment.
FIG. 6 is a diagram showing components of a system in one
embodiment that can automate feedback provision in language
usage.
FIG. 7 illustrates a schematic of an example computer or processing
system that may implement an automated language usage feedback
system in one embodiment.
FIG. 8 illustrates a cloud computing environment in one
embodiment.
FIG. 9 illustrates a set of functional abstraction layers provided
by cloud computing environment in one embodiment of the present
disclosure.
DETAILED DESCRIPTION
A system and method are disclosed for generating and/or providing
automated corrective feedback to users or students during written
language exercises taken on a computer. In one embodiment, the
system may include separately trained language models, for example,
one trained on "good" or "strong" language works (e.g., written
submissions determined to have proper language usage), the other on
"weak" language works (e.g., written submissions determined to have
weak or less proper language usage) of the same cohort. In one
embodiment, the system feeds in written work, a word at a time to
the language models. In one embodiment, the system determines the
deviation of the written work towards the weak language model. In
one embodiment, the system suggests or offers possible ways for the
writer to improve the work. A related method is also disclosed.
The system may improve an automated system and/or tool which
provides automatic corrections, for example, in language usage such
as written language. The system and/or method may provide guidance
in an automated manner to a user or student, for example, during
language exercises, for example, performed or taken on a
computer.
In one embodiment, the system may provide automatic, artificial
intelligence (AI)-driven guidance to a user during a written
language exercise. The system in one embodiment may extend
automatic corrections to include higher-level constructs, and for
example, can recognize incorrect usage of a language, for example,
incorrect word order and other aspects of sentence structure, usage
of articles, verb tenses, and/or others. The system in one
embodiment can be easier to deploy and have the property of being
able to use the written exercises from the system to train and
further improve itself over time.
The system, in one embodiment, which may be AI-based, can detect a
mistake in a language usage and provide a suggestion on how to
correct the mistake. The system can be made aware of colloquialism,
for example, and can offer guidance for avoiding it, if desired,
which may be difficult for a rules-based system to implement. The
system can model and then identify local or contextual mistakes
that differ from region to region. This guidance can suggest
changes moving text submissions from one state to another, for
example, from perceived "poor" to "good" quality or from one style
to another.
The system in one embodiment models the effects of linguist
background, and local and/or contextual language usage, for
example, to identify mistakes, which a generic tool may not be able
to identify. For example, the system in one embodiment can identify
the following sentence as an incorrectly structured sentence: "Also
me, I can be able to write an essay."
A language model (also known as (a.k.a.) a sequence model) is an
artificial intelligence construct (e.g., a Recurrent Neural
Network), which given a word or sequence of words, predicts the
next most likely word based on its training data. As an example, a
language model trained on a known author's works, when fed with the
word "just", might suggest that the next word is most likely
"after", and feeding those two words in (and so on for additional
outputs) might produce output that looks as follows (this output is
taken from an actual trained sequence model): "just after that
thereby how made with the way anything, and set for harmless
philos".
The above demonstrates that the language model has assimilated some
style, phrasing, and vocabulary from the training data. An
embodiment of a system and/or method in accordance with the present
disclosure, use multiple trained language models to detect mistakes
and to provide guidance. One or more language models are trained on
"good" or "strong" submissions from students, others on "weak"
submissions. The strong submissions language models may also be
trained on available content such as Wikipedia content, academic
papers or formal writings, newspaper or blog content, and/or
others. The weak submissions language models may be trained on
submissions of users or students from various grades and from
various schools, including schools from different locations or
countries. Strong and weak submissions may be obtained from other
or different sources.
Common local mistakes associated with a region, in writing the
English language, for example, can be observed on social networks,
such as social network blogs or messages or other content posted on
a social network website pages. In one embodiment, such local data
can be extracted, for example, based on specific hashtags that can
be attributed to certain groups of users. Social network blog data
can help the language model to learn local mistakes in language
usage, e.g., incorrect English usage, and also to evolve informal
writing that is used on social networks. Such a language model can
influence how learners write, for example, in a classroom or formal
setting. For instance, regional local information can be utilized
for correcting non-native language usage by learners in that region
learning to use (e.g., write) that non-native language. As an
example, local regional information associated with a learner of
the English language as a second language can be used by the
models.
In an example use case, a user can be entering written work on a
computer, for example, via a user interface. The system in one
embodiment can receive the user entry, for example, including words
forming a sentence. A plurality of such sentences can be received.
The system can extract a sentence, and incrementally feed the
sentence through the "weak" and "strong" ("good") submissions
language models, adding one additional word to the input at a time.
At each step the output that the model provides is a set of
predicted possible next words along with a score of how well the
predicted word fits into the language style of that model. The
system keeps track of a tally of the sentence scores for each
model.
Once the entire sentence has been fed through all of the models,
the system can have a score of how well the sentence matches to
each model, for instance, by comparing word by word a sentence
output by a model with the actual user written sentence. The system
can keep track of where the sentence deviated significantly from
the sentence output by one or more "good" or "strong" submissions
language models (e.g., trained based on submissions such as written
works determined to be strong or determined to have proper usage of
the language), which allows it to pinpoint an area or areas of the
sentence that need to be highlighted for the user or student to
review.
FIG. 1 is a diagram illustrating feeding of a sentence into two
models and processing of outputs of the models in one embodiment.
Consider the following sentence which deviates from well written
English in the final word: "You are excellent at something when you
do it good." The sentence 102 is input to a first model 104 and a
second model 106. The first model 104, for example, is a model
trained on well written works or works determined to be written
with proper usage of a given language. Given an input sentence or
phrase, the model trained on such data set can detect proper usage
of a language in the input sentence or phrase. An example of a
given language is the English language. The second model 106 is
trained on works or submissions that include weak use of the given
language, for example, those determined to use improper or awkward
use of the given language. Given an input sentence or phrase, the
model trained on such data set can detect weak or less proper usage
of the language in the given input sentence or phrase. The sentence
up until the current word is input to each of the language models.
The words in the sentence can be input to the first model 104 one
at a time. The words in the sentence can be input to the second
model 106 one at a time. Each of the two models 104, 106 can
independently predict a next word in the sentence, or a set of next
words with confidence scores. For example, the first model 104
predicts, based on its learned parameters (e.g., weights and bias),
the next word in the sentence. The first model 104 can also output
a confidence or likelihood score associated with the predicted next
word. For instance, as shown at 108, the first model 104 may
predict that the next word is "well" with confidence score of 90,
and also predict that the next word is "correctly" with confidence
score of 75. For example, the chance or probability of the next
word being "well" is 90; the chance or probability of the next word
being "correctly" is 75. Similarly, the second model 106 predicts,
based on its learned parameters (e.g., weights and bias), the next
word in the sentence. The second model 106 can also output a
confidence or likelihood score associated with the predicted next
word. For example, the output of each of the language models 104,
106 is a set of predicted next words along with their score
(likelihood). For instance, as shown at 110, the second model 108
may predict that the next word is "good" with likelihood score (or
probability) of 90, and also predict that the next word is "nice"
with likelihood score (or probability) of 85.
In this example, up until the final word of the sentence, there may
be similar scores predicted from the first language model 104
(e.g., strong submissions language model) and the second language
model 106 (e.g., weak submissions language model), for example, as
shown at 112. The word predictions along with the actual next word
and the current tally of strong and weak scores are passed into a
processor. Consider that the actual next word entered is "good".
The word predictions with their associated scores 108, 110, current
accumulation or tally 112 of scores from the first model 104 and
the second model 106, and the actual word 114, are input to a
processor 116. The processor 116, for example, can be a computer
executable component or module.
The processor 116 determines whether the actual next word (e.g.,
"good") appears in any of the first or second models' prediction
sets, e.g., "well" and "correctly" from the first model's output,
and "good" and "nice" from the second model's output. If the actual
word appears in the prediction set, the processor adds the word
score to the score tally associated with the model that predicted
the actual word. In this example the word "good" does not appear in
the first model's prediction set, so nothing is added to the tally
associated with the first model. However, the word "good" does
appear in the second model's prediction set, therefore, the word
score is added to the second model's tally. For example, as shown
at 118, the tally associated with the first model (e.g., strong
submissions language model) remains the same as the previous tally
(shown at 112); the tally associated with the second model (e.g.,
weak language model) is incremented by the score associated with
the word "good" (e.g., predicted by the weak submissions language
model with confidence score of "90", as shown at 110). The
processor 116 can keep track of the position of words where the
scores from the first model and second model (e.g., strong and weak
submissions language models) deviate. For example, deviation points
can include or specify a word position 10 (representing the
position of the word `good` in this example), which is the position
the two models deviated in their predictions.
Notification or alert with respect to words that deviate in the
predictions of the models 102, 104 can be communicated. For
example, words that deviate can be highlighted via a user
interface. In one embodiment, a system and/or method may cause,
direct or control a user interface or the like to highlight one or
more words at which the strong and weak submissions language models
probabilities deviate by more than a pre-selected or pre-defined
threshold. The pre-defined threshold can be configurable. FIG. 2
shows an example of a user interface display showing a sentence in
one embodiment. For instance, a user or student may be entering or
inputting (e.g., by typing in the sentence) on a computer via a
user interface such as a graphical user interface (GUI) on a
computer screen or another input mechanism, and the input sentence
can be presented or displayed on the computer screen. A word having
a probability deviation greater than the pre-defined threshold can
be highlighted via the user interface, for example, shown at 202.
Highlighting may be provided in any form of annotation, for
example, underline, shade, and/or others. The words on the user
interface can be rendered as a selectable object on the user
interface. In response to the user or student clicking on the
highlighted word, the user interface may show or present a set of
predictions 204 from the first model (e.g., strong submissions
language model) as a possible way to begin improving the sentence.
The set of predictions 204 can be presented as a pop-up object, a
call-out object, or another object on the user interface. This can
be viewed as being different from a correction, as no sentence
structure or phrase correction is suggested, but provides a
guidance that the student can follow to discover how to improve the
writing.
In another embodiment, language models can be trained in various
categories of works such as written works, such as but not limited
to: Shakespearean writing: trained on the works of William
Shakespeare Academic: trained on academic publications Formal:
trained on newspaper articles, Wikipedia, or other publications
Emotional: trained on emotive works of fiction Blog: trained on
numerous blog articles Poetry: trained on works of poetry
A user or student, for example, can choose a category of model to
focus the student's efforts on. As another example, a teacher or
instructor or the like can set a category for the students. Such a
category setting can diversify the user or student's writing
ability. In each category, a system can train both strong and weak
submissions language models. For example, a strong submissions
language model can be trained on the actual original corpus or
submissions of written work determined to have proper language
usage, which may be received, retrieved, or obtained from an
external source such as other schools, or departments related to
education for that category. A weak submissions language model can
be trained on previous submissions of students in that category
that were considered to include weak language usage.
In yet another embodiment, the submissions can be grouped into
cohorts, so that there are different models trained per cohort, for
each category. As an example, there may be models for "Poetry for
Grade 8" (one cohort), and models for "Poetry for Grade 10" (a
different cohort). Cohorts can also be based on different factors
or group categories. Another example of a cohort can be based on
proficiency level, for example "Poetry for [X] native speakers
taking English as a second language", wherein X can be a language
spoken in a given region or location. The cohort selection can be
made automatically by the system by taking into consideration the
context of the user (such as age, first language, location, etc.).
The system also may select a cohort automatically by comparing the
written work of the user to the various models and cohorts to
select the most similar one. This way, as the user improves in
proficiency, a different cohort can automatically be selected. In
the case of teacher grading, the system may cause, direct or
control a user interface to display a student's automatically
selected cohort, thus showing the teacher the proficiency level of
the student. The system can also display the deviations from the
strong submissions language model in order to assist the teacher in
grading.
In still yet another embodiment, a system may also compute a score
associated with language incorrectness of a written submission. In
one embodiment, such a score may not be displayed to the learner
unless the teacher decides to. The system can generate a report
associated with such a score per student and for the class. For
example, the system can show to the teacher the range of mistakes
that are common among the learners, and the progress of each
learner over time. A score associated with language incorrectness
can be a percentage value that indicates the number of words in the
student's submission that have been flagged with a high probability
of being incorrect or weak language (e.g., English) phrasing, with
reference to the total number of words in the submission. In one
embodiment, a model is a per-word model that checks the probability
of every word's closeness to strong or weak English usage, and the
number of corrected words indicates the extent to which the
submission deviates from the choice of words in proper English.
The system can collect, per student and for the class, the words
that are commonly corrected, and these words can be classified by
semantic similarity to understand the types of words that students
often get wrong. In addition, from these data (type of words), a
weak language model can further learn word sequences that have the
higher likelihood of being associated with incorrect language
usage.
A system and method in some embodiments can provide automated
guidance to students during language exercises (e.g., written
language exercise) entered electronically, for example, on a
computer. The guidance can highlight sentences (or words) in the
written work (for example, by underlining with a wavy line or by
another highlighting method) if the sentence correlates highly with
weak submissions. Whether the sentence correlates "highly" can be
determined based on comparing with a predefined threshold.
In some embodiments, the guidance can be based on using two or more
language models that have been trained on historical submissions.
The submissions can be obtained from various grades and various
schools. One group of language models can be trained on weak
submissions, while the other group can be trained on strong
submissions and/or strong examples of written work such as
publications. The system analyses the written work one sentence at
a time, by feeding in one word at a time into each set of language
models. The system can compare the outputs of the two groups of
language models against each other. If the sentence correlates more
closely to the language models trained on weak submissions than
that of the strong submissions, then the system can flag or cause
to flag the sentence or a deviating word, for example, by
highlighting or causing to highlight the sentence or the deviating
word on a user interface, for example, as needing review.
The correlation of the input sentence to the output of the language
models is calculated by feeding the sentence progressively into
each language model, and checking the probability of the next word
predicted by each language model, that the student wrote. For each
model (e.g., strong submissions and weak submissions language
models), the predicted probability per word can be stored, and
added up for the sentence. The model with the higher total tally
can be determined as the one the sentence most correlates with. One
or more words in the sentence can be highlighted or caused to be
highlighted, where the strong submissions language model and the
weak submissions language model's probabilities deviate by more
than a pre-selected threshold, for example, for review. Responsive
to the user clicking on or selecting the highlighted portion, one
or more alternate words can be presented or displayed. These
suggestions are the outputs of the language model(s) trained on
strong submissions, giving the student the opportunity to
incorporate into or rephrase their sentence.
In some embodiments, a "proficiency score" is calculated as a
percentage value that indicates the number of words in the
student's submission that have been flagged with a high probability
of incorrect usage or phrasing of the language (e.g., English), for
example, the number of words highlighted in comparison to the total
number of words in the submission.
FIG. 3 is an example model, which can be trained based on training
data set. As described herein, the training data set can include
data determined to have strong or proper language usage to train a
"good" or "strong" submissions language model. Another training
data set can include data determined to have weak language usage to
train a "weak" submissions language model. Further the training
data set can be grouped by different cohorts for different purposes
as described above.
In one embodiment, a model can be an artificial neural network
model, also referred to as a neural network model, for example,
shown at 300. An embodiment of an implementation of an artificial
neural network can include a succession of layers of neurons, which
are interconnected so that output signals of neurons in one layer
are weighted and transmitted to neurons in the next layer. A neuron
Ni in a given layer may be connected to one or more neurons Nj in
the next layer, and different weights wij can be associated with
each neuron-neuron connection Ni-Nj for weighting signals
transmitted from Ni to Nj. A neuron Nj generates output signals
dependent on its accumulated inputs, and weighted signals can be
propagated over successive layers of the network from an input to
an output neuron layer. An artificial neural network machine
learning model can undergo a training phase in which the sets of
weights associated with respective neuron layers are determined.
The network is exposed to a set of training data, in an iterative
training scheme in which the weights are repeatedly updated as the
network "learns" from the training data. The resulting trained
model, with weights defined via the training operation, can be
applied to perform a task based on new data.
An example of the neural network model is a recurrent neural
network model, which can handle time series data or sequence
based-data such as sentences in a language. A recurrent neural
network model can have a series of neural network cells 302a, 302b,
302n, which take as input a word in a sentence and also activation
information from the previous neural network in the previous time
step. For example, copies of neural network are made over time with
different inputs at different time steps. The copies of neural
network can share weights over time. The neural network at 302b can
take both the input word (e.g., x.sub.2) at that time step (e.g.,
t=2) and activation information from the previous neural network at
the previous time step (e.g., t=1), to predict the next word, e.g.,
its output y.sub.2. For example, the activation value a.sub.1 from
time step 1 is passed onto time step 2. The neural network 302b at
time step 2 uses both the activation value and input word to
predict the next word. Similarly, at time step n, the neural
network the next word in a given sentence is received along with
the activation value a.sub.2 computed at the previous time step to
predict its output (next word). At each time step, the recurrent
neural network passes on its activation to the next time step for
use. Thus, at the current step, both the input word for that step
and information from previous words in the sentence can be used to
predict the next word. In the figure, a.sub.0 can be an initial
activation vector, which can be initialized to zeros or other
initial values. A language model can predict a next word, for
example, given a word in a sentence, for example, in sequence. For
instance, such a language model may output a percentage value or
score that the next word is word xyz.
The architecture shown in FIG. 3 is only an example of a neural
network, an example of a recurrent neural network model, which can
be used to generate the language models described herein. Other and
different model architecture can be used. For example, different
types of recurrent neural network models such as but not limited to
Long Short-Term Memory (LSTM), different types of neural network
models, different types of deep learning neural network models can
be used. Yet in other aspects, different types of artificial
intelligence models can be used.
FIG. 4 is a flow diagram illustrating a method in one embodiment,
which can provide automated corrective feed. At 402, the method may
include inputting an input set of words in a given sentence to a
first language model. The first language model outputs a first set
of candidate words predicted to follow the input set of words in
the given sentence. The first language model further outputs a
score associated with each of the candidate words in the first set
of candidate words.
At 404, the method may also include inputting the input set of
words in the given sentence to a second language model. The second
language model output a second set of candidate words predicted to
follow the input set of words in the given sentence. The second
language model further outputs a score associated with each of the
candidate words in the second set of candidate words. In some
embodiments, the given sentence can be fed into the first language
model and the second language model one word at a time.
At 406, the method may include receiving an actual word following
the input set of words. For instance, a user may have entered the
next word via a user interface following the input set of words,
and that next word is received as the actual word. The actual word
may be compared with the output of the first language model, and
the output of the second language model.
At 408, responsive to determining that the actual word matches with
a candidate word in the first set of candidate words, the method
may include updating a first cumulative tally associated with the
first language model with the score associated with the candidate
word in the first set matching the actual word.
At 410, responsive to determining that the actual word matches with
a candidate word in the second set of candidate words, the method
may include updating a second cumulative tally associated with the
second language model with the score associated with the candidate
word in the second set matching the actual word.
The first cumulative tally and the second cumulative tally may be
compared. At 412, responsive to determining that the first
cumulative tally and the second cumulative tally deviate by more
than a pre-defined threshold, the actual word in the given sentence
can be flagged. For example, a user interface can be caused to
display or highlight the actual word in the given sentence. In some
embodiments, the method may further include causing a presentation
of the first set of candidate words responsive to the flagged
actual word being selected.
In some embodiments, a method may also include training the first
language model based on a first training data set. The first
language model is trained to predict a first set of next words to
follow an input set of words in a given sentence. The first
language model can be trained to output the first set of next words
and a score associated with each of the next words in the first
set.
In some embodiments, a method may also include training the second
language model based on a second training data set. The second
language model is trained to predict a second set of next words to
follow an input set of words in the given sentence. The second
language model can be trained to output the second set of next
words and a score associated with each of the next words in the
second set.
The first language model and the second language model can be
artificial neural network models, for example, recurrent neural
network models, but not limited to only those types of models.
Other types of models can be trained. The first language model can
be trained based on a first training data set including at least
data determined to have proper usage of a language, and the second
language model can be trained based on a second training data set
including at least data determined to have incorrect usage of the
language.
In some embodiments, the first training data set can include at
least data associated with a category of submission. The first
language model can be trained to output the first set of next words
likely to be associated with language found in the category of
submission.
In some embodiments, the first training data set can include at
least data associated with a category of submission grouped by a
cohort. The first language model can be trained to output the first
set of next words likely to be associated with language found in
the category of submission grouped by the cohort.
FIG. 5 is a flow diagram illustrating a method in another
embodiment. At 502, the method may include running a first language
model with an input set of words in a given sentence. The first
language model outputs a first set of candidate words predicted to
follow the input set of words in the given sentence. At 504, the
method may include running a first language model with the input
set of words in the given sentence. The second language model
outputs a second set of candidate words predicted to follow the
input set of words in the given sentence. At 506, based on the
first set of candidate words, the second set of candidate words,
and an actual next word following the input set of words, the
method may include providing guidance for phrasing the given
sentence.
FIG. 6 is a diagram showing components of a system in one
embodiment that can automate feedback provision in language usage.
One or more hardware processors 602 such as a central processing
unit (CPU), a graphic process unit (GPU), and/or a Field
Programmable Gate Array (FPGA), an application specific integrated
circuit (ASIC), and/or another processor, may be coupled with a
memory device 604, and provide automate feedback in language usage,
for example, during a language exercise. A memory device 604 may
include random access memory (RAM), read-only memory (ROM) or
another memory device, and may store data and/or processor
instructions for implementing various functionalities associated
with the methods and/or systems described herein.
One or more processors 602 may execute computer instructions stored
in memory 604 or received from another computer device or medium. A
memory device 604 may, for example, store instructions and/or data
for functioning of one or more hardware processors 602, and may
include an operating system and other program of instructions
and/or data. The memory device 604 may also store a first language
model and a second language model.
One or more hardware processors 602 may run the first language
model with an input set of words in a given sentence. The first
language model may output a first set of candidate words predicted
to follow the input set of words in the given sentence. The first
language model may further output a score associated with each of
the candidate words in the first set of candidate words. One or
more hardware processors 602 may also run the second language model
with the input set of words in the given sentence. The second
language model may output a second set of candidate words predicted
to follow the input set of words in the given sentence. The second
language model may further output a score associated with each of
the candidate words in the second set of candidate words. A given
sentence can be fed into the first language model and the second
language model one word at a time.
One or more hardware processors 602 also may receive an actual word
following the input set of words. For example, a user may have
written or entered the next word following the input set of words.
One or more hardware processors 602, responsive to determining that
the actual word matches with a candidate word in the first set of
candidate words, may update a first cumulative tally associated
with the first language model with the score associated with the
candidate word in the first set matching the actual word. One or
more hardware processors 602, responsive to determining that the
actual word matches with a candidate word in the second set of
candidate words, may update a second cumulative tally associated
with the second language model with the score associated with the
candidate word in the second set matching the actual word. One or
more hardware processors 602, responsive to determining that the
first cumulative tally and the second cumulative tally deviate by
more than a pre-defined threshold, identify the actual word in the
given sentence for flagging.
In some embodiments, one or more hardware processors 602 may also
train the first language model and the second language model. The
first language model may be trained based on a first training data
set. The first language model is trained to predict a first set of
next words to follow an input set of words in a given sentence. The
first language model can be trained to output the first set of next
words and a score associated with each of the next words in the
first set. The second language model may be trained based on a
second training data set. The second language model is trained to
predict a second set of next words to follow an input set of words
in the given sentence. The second language model can be trained to
output the second set of next words and a score associated with
each of the next words in the second set.
In some embodiments, the first language model and the second
language model are artificial neural network models. For example,
the first language model and the second language model can be
recurrent neural network models.
The first language model can be trained based on a first training
data set including at least data determined to have proper usage of
a language, and the second language model can be trained based on a
second training data set including at least data determined to have
incorrect usage of the language.
In some embodiments, the first training data set can include at
least data associated with a category of submission, and the first
language model can be trained to output the first set of next words
likely to be associated with language found in the category of
submission.
In some embodiments, the first training data set can include at
least data associated with a category of submission grouped by a
cohort, and the first language model can be trained to output the
first set of next words likely to be associated with language found
in the category of submission grouped by the cohort.
In some embodiments, the first and/or second training data sets may
be stored in a storage device 606 or received or obtained via a
network interface 608 from a remote device, and may be temporarily
loaded into a memory device 604 for building or generating the
models. The learned models may be stored on a memory device 604,
for example, for execution by one or more hardware processors 602.
One or more hardware processors 602 may be coupled with interface
devices such as a network interface 608 for communicating with
remote systems, for example, via a network, and an input/output
interface 610 for communicating with input and/or output devices
such as a keyboard, mouse, display, and/or others.
In some embodiments, one or more hardware processors 602 can cause
flagging of the actual word in the given sentence. One or more
hardware processors 602 can also cause highlighting of the actual
word in the given sentence. One or more hardware processors 602 can
provide the first set of candidate words, and for example, cause a
presentation of the first set of candidate words, for example, via
a user interface, for example, to be displayed in the vicinity of
the highlighted portion of the given sentence.
Unless otherwise explicitly noted, one or more elements, features
and/or components of various embodiments described herein can be
mixed and/or combined.
FIG. 7 illustrates a schematic of an example computer or processing
system that may implement an automated language usage feedback
system in one embodiment of the present disclosure. The computer
system is only one example of a suitable processing system and is
not intended to suggest any limitation as to the scope of use or
functionality of embodiments of the methodology described herein.
The processing system shown may be operational with numerous other
general purpose or special purpose computing system environments or
configurations. Examples of well-known computing systems,
environments, and/or configurations that may be suitable for use
with the processing system shown in FIG. 7 may include, but are not
limited to, personal computer systems, server computer systems,
thin clients, thick clients, handheld or laptop devices,
multiprocessor systems, microprocessor-based systems, set top
boxes, programmable consumer electronics, network PCs, minicomputer
systems, mainframe computer systems, and distributed cloud
computing environments that include any of the above systems or
devices, and the like.
The computer system may be described in the general context of
computer system executable instructions, such as program modules,
being executed by a computer system. Generally, program modules may
include routines, programs, objects, components, logic, data
structures, and so on that perform particular tasks or implement
particular abstract data types. The computer system may be
practiced in distributed cloud computing environments where tasks
are performed by remote processing devices that are linked through
a communications network. In a distributed cloud computing
environment, program modules may be located in both local and
remote computer system storage media including memory storage
devices.
The components of computer system may include, but are not limited
to, one or more processors or processing units 12, a system memory
16, and a bus 14 that couples various system components including
system memory 16 to processor 12. The processor 12 may include a
module 30 that performs the methods described herein. The module 30
may be programmed into the integrated circuits of the processor 12,
or loaded from memory 16, storage device 18, or network 24 or
combinations thereof.
Bus 14 may represent one or more of any of several types of bus
structures, including a memory bus or memory controller, a
peripheral bus, an accelerated graphics port, and a processor or
local bus using any of a variety of bus architectures. By way of
example, and not limitation, such architectures include Industry
Standard Architecture (ISA) bus, Micro Channel Architecture (MCA)
bus, Enhanced ISA (EISA) bus, Video Electronics Standards
Association (VESA) local bus, and Peripheral Component
Interconnects (PCI) bus.
Computer system may include a variety of computer system readable
media. Such media may be any available media that is accessible by
computer system, and it may include both volatile and non-volatile
media, removable and non-removable media.
System memory 16 can include computer system readable media in the
form of volatile memory, such as random access memory (RAM) and/or
cache memory or others. Computer system may further include other
removable/non-removable, volatile/non-volatile computer system
storage media. By way of example only, storage system 18 can be
provided for reading from and writing to a non-removable,
non-volatile magnetic media (e.g., a "hard drive"). Although not
shown, a magnetic disk drive for reading from and writing to a
removable, non-volatile magnetic disk (e.g., a "floppy disk"), and
an optical disk drive for reading from or writing to a removable,
non-volatile optical disk such as a CD-ROM, DVD-ROM or other
optical media can be provided. In such instances, each can be
connected to bus 14 by one or more data media interfaces.
Computer system may also communicate with one or more external
devices 26 such as a keyboard, a pointing device, a display 28,
etc.; one or more devices that enable a user to interact with
computer system; and/or any devices (e.g., network card, modem,
etc.) that enable computer system to communicate with one or more
other computing devices. Such communication can occur via
Input/Output (I/O) interfaces 20.
Still yet, computer system can communicate with one or more
networks 24 such as a local area network (LAN), a general wide area
network (WAN), and/or a public network (e.g., the Internet) via
network adapter 22. As depicted, network adapter 22 communicates
with the other components of computer system via bus 14. It should
be understood that although not shown, other hardware and/or
software components could be used in conjunction with computer
system. Examples include, but are not limited to: microcode, device
drivers, redundant processing units, external disk drive arrays,
RAID systems, tape drives, and data archival storage systems,
etc.
It is understood in advance that although this disclosure may
include a description on cloud computing, implementation of the
teachings recited herein are not limited to a cloud computing
environment. Rather, embodiments of the present invention are
capable of being implemented in conjunction with any other type of
computing environment now known or later developed. Cloud computing
is a model of service delivery for enabling convenient, on-demand
network access to a shared pool of configurable computing resources
(e.g. networks, network bandwidth, servers, processing, memory,
storage, applications, virtual machines, and services) that can be
rapidly provisioned and released with minimal management effort or
interaction with a provider of the service. This cloud model may
include at least five characteristics, at least three service
models, and at least four deployment models.
Characteristics are as Follows:
On-demand self-service: a cloud consumer can unilaterally provision
computing capabilities, such as server time and network storage, as
needed automatically without requiring human interaction with the
service's provider.
Broad network access: capabilities are available over a network and
accessed through standard mechanisms that promote use by
heterogeneous thin or thick client platforms (e.g., mobile phones,
laptops, and PDAs).
Resource pooling: the provider's computing resources are pooled to
serve multiple consumers using a multi-tenant model, with different
physical and virtual resources dynamically assigned and reassigned
according to demand. There is a sense of location independence in
that the consumer generally has no control or knowledge over the
exact location of the provided resources but may be able to specify
location at a higher level of abstraction (e.g., country, state, or
datacenter).
Rapid elasticity: capabilities can be rapidly and elastically
provisioned, in some cases automatically, to quickly scale out and
rapidly released to quickly scale in. To the consumer, the
capabilities available for provisioning often appear to be
unlimited and can be purchased in any quantity at any time.
Measured service: cloud systems automatically control and optimize
resource use by leveraging a metering capability at some level of
abstraction appropriate to the type of service (e.g., storage,
processing, bandwidth, and active user accounts). Resource usage
can be monitored, controlled, and reported providing transparency
for both the provider and consumer of the utilized service.
Service Models are as Follows:
Software as a Service (SaaS): the capability provided to the
consumer is to use the provider's applications running on a cloud
infrastructure. The applications are accessible from various client
devices through a thin client interface such as a web browser
(e.g., web-based e-mail). The consumer does not manage or control
the underlying cloud infrastructure including network, servers,
operating systems, storage, or even individual application
capabilities, with the possible exception of limited user-specific
application configuration settings.
Platform as a Service (PaaS): the capability provided to the
consumer is to deploy onto the cloud infrastructure
consumer-created or acquired applications created using programming
languages and tools supported by the provider. The consumer does
not manage or control the underlying cloud infrastructure including
networks, servers, operating systems, or storage, but has control
over the deployed applications and possibly application hosting
environment configurations.
Infrastructure as a Service (IaaS): the capability provided to the
consumer is to provision processing, storage, networks, and other
fundamental computing resources where the consumer is able to
deploy and run arbitrary software, which can include operating
systems and applications. The consumer does not manage or control
the underlying cloud infrastructure but has control over operating
systems, storage, deployed applications, and possibly limited
control of select networking components (e.g., host firewalls).
Deployment Models are as Follows:
Private cloud: the cloud infrastructure is operated solely for an
organization. It may be managed by the organization or a third
party and may exist on-premises or off-premises.
Community cloud: the cloud infrastructure is shared by several
organizations and supports a specific community that has shared
concerns (e.g., mission, security requirements, policy, and
compliance considerations). It may be managed by the organizations
or a third party and may exist on-premises or off-premises.
Public cloud: the cloud infrastructure is made available to the
general public or a large industry group and is owned by an
organization selling cloud services.
Hybrid cloud: the cloud infrastructure is a composition of two or
more clouds (private, community, or public) that remain unique
entities but are bound together by standardized or proprietary
technology that enables data and application portability (e.g.,
cloud bursting for load-balancing between clouds).
A cloud computing environment is service oriented with a focus on
statelessness, low coupling, modularity, and semantic
interoperability. At the heart of cloud computing is an
infrastructure that includes a network of interconnected nodes.
Referring now to FIG. 8, illustrative cloud computing environment
50 is depicted. As shown, cloud computing environment 50 includes
one or more cloud computing nodes 10 with which local computing
devices used by cloud consumers, such as, for example, personal
digital assistant (PDA) or cellular telephone 54A, desktop computer
54B, laptop computer 54C, and/or automobile computer system 54N may
communicate. Nodes 10 may communicate with one another. They may be
grouped (not shown) physically or virtually, in one or more
networks, such as Private, Community, Public, or Hybrid clouds as
described hereinabove, or a combination thereof. This allows cloud
computing environment 50 to offer infrastructure, platforms and/or
software as services for which a cloud consumer does not need to
maintain resources on a local computing device. It is understood
that the types of computing devices 54A-N shown in FIG. 8 are
intended to be illustrative only and that computing nodes 10 and
cloud computing environment 50 can communicate with any type of
computerized device over any type of network and/or network
addressable connection (e.g., using a web browser).
Referring now to FIG. 9, a set of functional abstraction layers
provided by cloud computing environment 50 (FIG. 8) is shown. It
should be understood in advance that the components, layers, and
functions shown in FIG. 9 are intended to be illustrative only and
embodiments of the invention are not limited thereto. As depicted,
the following layers and corresponding functions are provided:
Hardware and software layer 60 includes hardware and software
components. Examples of hardware components include: mainframes 61;
RISC (Reduced Instruction Set Computer) architecture based servers
62; servers 63; blade servers 64; storage devices 65; and networks
and networking components 66. In some embodiments, software
components include network application server software 67 and
database software 68.
Virtualization layer 70 provides an abstraction layer from which
the following examples of virtual entities may be provided: virtual
servers 71; virtual storage 72; virtual networks 73, including
virtual private networks; virtual applications and operating
systems 74; and virtual clients 75.
In one example, management layer 80 may provide the functions
described below. Resource provisioning 81 provides dynamic
procurement of computing resources and other resources that are
utilized to perform tasks within the cloud computing environment.
Metering and Pricing 82 provide cost tracking as resources are
utilized within the cloud computing environment, and billing or
invoicing for consumption of these resources. In one example, these
resources may include application software licenses. Security
provides identity verification for cloud consumers and tasks, as
well as protection for data and other resources. User portal 83
provides access to the cloud computing environment for consumers
and system administrators. Service level management 84 provides
cloud computing resource allocation and management such that
required service levels are met. Service Level Agreement (SLA)
planning and fulfillment 85 provide pre-arrangement for, and
procurement of, cloud computing resources for which a future
requirement is anticipated in accordance with an SLA.
Workloads layer 90 provides examples of functionality for which the
cloud computing environment may be utilized. Examples of workloads
and functions which may be provided from this layer include:
mapping and navigation 91; software development and lifecycle
management 92; virtual classroom education delivery 93; data
analytics processing 94; transaction processing 95; and automated
feedback processing 96.
The present invention may be a system, a method, and/or a computer
program product at any possible technical detail level of
integration. The computer program product may include a computer
readable storage medium (or media) having computer readable program
instructions thereon for causing a processor to carry out aspects
of the present invention.
The computer readable storage medium can be a tangible device that
can retain and store instructions for use by an instruction
execution device. The computer readable storage medium may be, for
example, but is not limited to, an electronic storage device, a
magnetic storage device, an optical storage device, an
electromagnetic storage device, a semiconductor storage device, or
any suitable combination of the foregoing. A non-exhaustive list of
more specific examples of the computer readable storage medium
includes the following: a portable computer diskette, a hard disk,
a random access memory (RAM), a read-only memory (ROM), an erasable
programmable read-only memory (EPROM or Flash memory), a static
random access memory (SRAM), a portable compact disc read-only
memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a
floppy disk, a mechanically encoded device such as punch-cards or
raised structures in a groove having instructions recorded thereon,
and any suitable combination of the foregoing. A computer readable
storage medium, as used herein, is not to be construed as being
transitory signals per se, such as radio waves or other freely
propagating electromagnetic waves, electromagnetic waves
propagating through a waveguide or other transmission media (e.g.,
light pulses passing through a fiber-optic cable), or electrical
signals transmitted through a wire.
Computer readable program instructions described herein can be
downloaded to respective computing/processing devices from a
computer readable storage medium or to an external computer or
external storage device via a network, for example, the Internet, a
local area network, a wide area network and/or a wireless network.
The network may comprise copper transmission cables, optical
transmission fibers, wireless transmission, routers, firewalls,
switches, gateway computers and/or edge servers. A network adapter
card or network interface in each computing/processing device
receives computer readable program instructions from the network
and forwards the computer readable program instructions for storage
in a computer readable storage medium within the respective
computing/processing device.
Computer readable program instructions for carrying out operations
of the present invention may be assembler instructions,
instruction-set-architecture (ISA) instructions, machine
instructions, machine dependent instructions, microcode, firmware
instructions, state-setting data, configuration data for integrated
circuitry, or either source code or object code written in any
combination of one or more programming languages, including an
object oriented programming language such as Smalltalk, C++, or the
like, and procedural programming languages, such as the "C"
programming language or similar programming languages. The computer
readable program instructions may execute entirely on the user's
computer, partly on the user's computer, as a stand-alone software
package, partly on the user's computer and partly on a remote
computer or entirely on the remote computer or server. In the
latter scenario, the remote computer may be connected to the user's
computer through any type of network, including a local area
network (LAN) or a wide area network (WAN), or the connection may
be made to an external computer (for example, through the Internet
using an Internet Service Provider). In some embodiments,
electronic circuitry including, for example, programmable logic
circuitry, field-programmable gate arrays (FPGA), or programmable
logic arrays (PLA) may execute the computer readable program
instructions by utilizing state information of the computer
readable program instructions to personalize the electronic
circuitry, in order to perform aspects of the present
invention.
Aspects of the present invention are described herein with
reference to flowchart illustrations and/or block diagrams of
methods, apparatus (systems), and computer program products
according to embodiments of the invention. It will be understood
that each block of the flowchart illustrations and/or block
diagrams, and combinations of blocks in the flowchart illustrations
and/or block diagrams, can be implemented by computer readable
program instructions.
These computer readable program instructions may be provided to a
processor of a general purpose computer, special purpose computer,
or other programmable data processing apparatus to produce a
machine, such that the instructions, which execute via the
processor of the computer or other programmable data processing
apparatus, create means for implementing the functions/acts
specified in the flowchart and/or block diagram block or blocks.
These computer readable program instructions may also be stored in
a computer readable storage medium that can direct a computer, a
programmable data processing apparatus, and/or other devices to
function in a particular manner, such that the computer readable
storage medium having instructions stored therein comprises an
article of manufacture including instructions which implement
aspects of the function/act specified in the flowchart and/or block
diagram block or blocks.
The computer readable program instructions may also be loaded onto
a computer, other programmable data processing apparatus, or other
device to cause a series of operational steps to be performed on
the computer, other programmable apparatus or other device to
produce a computer implemented process, such that the instructions
which execute on the computer, other programmable apparatus, or
other device implement the functions/acts specified in the
flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the Figures illustrate the
architecture, functionality, and operation of possible
implementations of systems, methods, and computer program products
according to various embodiments of the present invention. In this
regard, each block in the flowchart or block diagrams may represent
a module, segment, or portion of instructions, which comprises one
or more executable instructions for implementing the specified
logical function(s). In some alternative implementations, the
functions noted in the blocks may occur out of the order noted in
the Figures. For example, two blocks shown in succession may, in
fact, be executed substantially concurrently, or the blocks may
sometimes be executed in the reverse order, depending upon the
functionality involved. It will also be noted that each block of
the block diagrams and/or flowchart illustration, and combinations
of blocks in the block diagrams and/or flowchart illustration, can
be implemented by special purpose hardware-based systems that
perform the specified functions or acts or carry out combinations
of special purpose hardware and computer instructions.
The terminology used herein is for the purpose of describing
particular embodiments only and is not intended to be limiting of
the invention. As used herein, the singular forms "a", "an" and
"the" are intended to include the plural forms as well, unless the
context clearly indicates otherwise. It will be further understood
that the terms "comprise", "comprises", "comprising", "include",
"includes", "including", and/or "having," when used herein, can
specify the presence of stated features, integers, steps,
operations, elements, and/or components, but do not preclude the
presence or addition of one or more other features, integers,
steps, operations, elements, components, and/or groups thereof.
The corresponding structures, materials, acts, and equivalents of
all means or step plus function elements, if any, in the claims
below are intended to include any structure, material, or act for
performing the function in combination with other claimed elements
as specifically claimed. The description of the present invention
has been presented for purposes of illustration and description,
but is not intended to be exhaustive or limited to the invention in
the form disclosed. Many modifications and variations will be
apparent to those of ordinary skill in the art without departing
from the scope and spirit of the invention. The embodiment was
chosen and described in order to best explain the principles of the
invention and the practical application, and to enable others of
ordinary skill in the art to understand the invention for various
embodiments with various modifications as are suited to the
particular use contemplated. Embodiments and/or components of
embodiments disclosed herein can be freely combined with each other
if they are not mutually exclusive.
* * * * *
References